Skip to content
Draft
Show file tree
Hide file tree
Changes from all commits
Commits
File filter

Filter by extension

Filter by extension

Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
2 changes: 2 additions & 0 deletions flash/apps/requests.mdx
Original file line number Diff line number Diff line change
Expand Up @@ -85,6 +85,8 @@ curl -X POST https://api.runpod.ai/v2/abc123xyz/runsync \
}
```

The `/runsync` endpoint has a 60-second client-side timeout by default. If you've configured `execution_timeout_ms` on your endpoint, the client timeout uses that value instead. For jobs that take longer than 60 seconds, set `execution_timeout_ms` to prevent `/runsync` requests from timing out.
Copy link
Contributor Author

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

Citation: PR #272 fixes ServerlessResource.runsync() to respect executionTimeoutMs for client-side HTTP timeout. The PR description explains the prior bug: 60s hardcoded timeout caused GPU inference jobs >1 minute to fail.
View source


<Tip>
Use `/run` for long-running jobs that you'll check later. Use `/runsync` for quick jobs where you want immediate results (with timeout protection).
</Tip>
Expand Down
4 changes: 4 additions & 0 deletions flash/configuration/parameters.mdx
Original file line number Diff line number Diff line change
Expand Up @@ -346,6 +346,10 @@ async def train(data): ...
async def infer(data): ...
```

<Note>
The Flash SDK's `runsync()` method uses your `execution_timeout_ms` value as the client-side HTTP timeout. If set to a positive value, the SDK waits that duration for the job to complete. If unset or set to `0`, the SDK defaults to a 60-second timeout. For long-running inference jobs, set `execution_timeout_ms` to prevent premature timeouts.
Copy link
Contributor Author

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

Citation: Bug fix PR that changed runsync() to use executionTimeoutMs / 1000 as client timeout instead of hardcoded 60s. Changes in src/runpod_flash/core/resources/serverless.py show the timeout derivation logic with guard for None and 0 values.
View source

</Note>

### flashboot

**Type**: `bool`
Expand Down
Loading